Advance Publication by-J-STAGE Circulation Journal REVIEW Official Journal of the Japanese Circulation Society http://www.j-circ.or.jp Biochemical and Physiological Regulation of Cardiac Myocyte Contraction by Cardiac-Specific Light Chain Kinase Osamu Tsukamoto, MD, PhD; Masafumi Kitakaze, MD, PhD

Cardiac-specific myosin light chain kinase (cMLCK) is the kinase predominantly responsible for the maintenance of the basal level of phosphorylation of cardiac myosin light chain 2 (MLC2), which it phosphorylates at Ser-15. This phosphorylation repels the myosin heads from the thick myosin filament and moves them toward the thin fila- ment. Unlike smooth muscle cells, MLC2 phosphorylation in striated muscle cells appears to be a positive modula- tor of Ca2+ sensitivity that shifts the Ca2+-force relationship toward the left and increases the maximal force response and thus does not initiate muscle contraction. Recent studies have revealed an increasing number of details of the biochemical, physiological, and pathophysiological characteristics of cMLCK. The combination of recent techno- logical advances and the discovery of a novel class of biologically active nonstandard peptides will hopefully translate into the development of drugs for the treatment of heart diseases.

Key Words: Ca2+/; Muscle contraction; Myosin light chain kinase; Regulatory myosin light chain

ctin and myosin are essential for the generation of con- smooth muscle contraction still remains unknown. tractile force. In 1954, Huxley and Hanson proposed Following the discovery of smMLCK and skeletal muscle A the “sliding filament theory”, which states that con- MLCK (skMLCK), a third MLCK, named cardiac MLCK tractile force is generated through slippage of the cross-bridg- (cMLCK), was identified in 20077 and is expressed exclusively es between actin and myosin filaments1 and that purified actin in cardiac myocytes. However, what is the role of MLCK in and myosin are not sufficient for active movement. In 1962, striated muscle cells? Because both skeletal and cardiac muscle intracellular calcium ([Ca2+]i) was identified as acting a trigger cells express specific MLCKs (ie, skMLCK and cMLCK, re- of muscle contraction.2 Although force generation can be spectively), MLCKs might play specific physiological roles in achieved with only myosin and actin, additional are each cell type. required for its regulation. In smooth muscle, an increase in 2+ 2+ [Ca ]i induces the binding of Ca /calmodulin to smooth mus- Intracellular Ca2+ Homeostasis in cle-specific myosin light chain kinase (smMLCK), which leads to the phosphorylation of cardiac myosin light chain 2 (MLC2) Muscle Tissues in thick myosin filaments, activation of the ATPase of myo- To set the stage for our discussion of MLCKs, it is necessary sin heads and initiation of muscle contraction.3 Thus, smooth to understand the intracellular Ca2+ homeostasis in muscle tis- muscle cells use smMLCK as a [Ca2+]i sensor and initiator of sues, because Ca2+ is a trigger of MLCK activation and the muscle contraction4 (Figure 1). In contrast, is spe- responses to [Ca2+]i differs among the 3 types of muscle cell: cifically expressed in skeletal and cardiac muscle cells and smooth muscle, skeletal muscle, and cardiac muscle (Figure 1). works as a sensing of [Ca2+]i.5 In contrast to smooth Local [Ca2+]i transients (Ca2+ sparks), which arise from the muscle cells, the initiation of striated muscle contraction is coordinated opening of a cluster of ryanodine-sensitive Ca2+ regulated primarily by the troponin- complex in release channels, are necessary for effective Ca2+ signaling in thin actin filaments and not through MLC2 phosphorylation. all 3 types of muscle cell, although the rates of Ca2+ transients The binding of Ca2+ to leads to relief of the inhibi- and contraction vary considerably in the different muscle cells. tion of the binding of cross-bridges to thin actin filaments, In general, however, these rates are slower in smooth muscle and this relief triggers striated muscle contraction (Figure 1). cells than in either skeletal or cardiac muscle cells. Smooth However, endogenous expression of troponin proteins in muscle cells gradually contract according to this slow increase smooth muscle cells was reported recently,6 although the phys- in the level of [Ca2+]i.8 In skeletal muscle cells, the activation iological role of the troponin complex in the regulation of of the voltage-gated Ca2+ channel in the transverse tubule mem-

Received May 20, 2013; revised manuscript received June 7, 2013; accepted June 17, 2013; released online July 18, 2013 Department of Molecular Cardiology, Osaka University Graduate School of Medicine, Suita (O.T.); Department of Cardiovascular Medicine, National Cerebral and Cardiovascular Center, Suita (M.K.), Japan Mailing address: Osamu Tsukamoto, MD, PhD, Department of Molecular Cardiology, Osaka University Graduate School of Medicine, 2-2 Yamadaoka, Suita 565-0871, Japan. E-mail: [email protected] ISSN-1346-9843 doi: 10.1253/circj.CJ-13-0627 All rights are reserved to the Japanese Circulation Society. For permissions, please e-mail: [email protected] Advance Publication by-J-STAGE TSUKAMOTO O et al.

Figure 1. Intracellular calcium dynamics and muscle contraction. CaM, calmodulin; DAG, diacylglycerol; DHPR, dihydropyridine receptor; ELC, essential myosin light chain; GPCR, G protein-coupled receptor; IP3, inositol trisphosphate; IP3R, IP3 receptor; MLCK, myosin light chain kinase; NCX, sodium-calcium exchanger; PLC-b, phospholipase C beta; PI(4,5)P2, phosphatidylinositol 4,5-bisphosphate; PMCA, plasma membrane Ca2+-ATPase; SERCA, sarco/endoplasmic reticulum Ca2+-ATPase; SR, sarcoplasmic reticulum; RLC, regulatory myosin light chain; RyR, ryanodine receptor; TnC, troponin C; TnI, ; TnT, ; T-tube, transverse tubules. For explanations, see text.

2+ branes directly opens the ryanodine-sensitive Ca release re- Mechanism of MLCK Activation by ceptor on the sarcoplasmic reticulum (SR), which results in a 2+ rapid increase in [Ca2+]i. In cardiac muscle cells, electrostimu- Ca /Calmodulin Binding lation from the sinus node activates the voltage-dependent An increase in [Ca2+]i can enhance MLCK activity approxi- cation channel and thus increases the [Ca2+]i level, which in- mately 1,000-fold through binding to calmodulin. Ca2+/calmod- duces a rapid and large release of Ca2+ from the SR, known as ulin is the most important regulator of MLCKs. MLCK is “Ca2+-induced Ca2+ release”. Importantly, the duration of the catalytically inactive in the absence of Ca2+/calmodulin be- increased [Ca2+]i level in striated muscle cells is very short cause the autoinhibitory sequence of MLCK blocks the access because the released Ca2+ is rapidly recaptured in the SR. In of the substrate to the catalytic core (Figure 2).9 The binding cardiac myocytes, electrostimulation increases [Ca2+]i from of Ca2+/calmodulin to the calmodulin-binding domain displac- 100 nmol/L to a few hundred nmol/L (~1 μmol/L), which re- es the autoinhibitory sequence from the surface of the catalytic turns to the basal level within 0.1 s. core, which results in exposure of the catalytic site of the ki- We now attempt to show why striated muscle cells do not nase and thus provides access to the N-terminus of MLC2.9 use MLCK as a sensor of [Ca2+]i and an initiator of muscle MLCK is maximally activated by Ca2+/calmodulin at a molar contraction. The slow kinase reactions of MLCKs are not ratio of 1:1 with a dissociation constant of 1 nmol/L.10 suitable as a [Ca2+]i sensor in striated muscles,4 in which rapid and transient increase in the [Ca2+]i occurs. Instead, using a non-enzyme signal from the binding of Ca2+ to tro- of MLCKs ponin, these cells contract promptly in response to the in- smMLCK creased [Ca2+]i. Thus, the increase in [Ca2+]i during muscle smMLCK is encoded by the single-copy MYLK1 gene, which contraction exhibits distinct molecular kinetics in the 3 types expresses 3 transcripts in a cell-specific manner related to al- of muscle, and these kinetics are related to the different ternative initiation sites: non-muscle isoform (longer form), physiological mechanisms that regulate contraction in these smooth muscle isoform (shorter form), and telokin.11 Nor- muscles. mally, smooth muscle expresses the shorter form of smMLCK. The smooth muscle isoform contains 3 DFRxxL motifs, a Advance Publication by-J-STAGE Cardiac-Specific MLCK

Figure 2. Structural and functional elements in myosin light chain kinase (MLCK) proteins. C, cardiac; CaM, calmodulin; CD, catalytic domain; Fn, fibronectin domain; Ig, immunoglobulin domain; RD, regulatory domain; sk, skeletal; sm, smooth muscle;. For explanations, see text.

proline-rich repeat, 3 immunoglobulin (Ig) modules, 1 fibro- skMLCK nectin (Fn) module, and 1 kinase domain with a catalytic core skMLCK is encoded by the MYLK2 gene and is predomi- and a regulatory segment. The 3 DFRxxL motifs at the N-ter- nantly expressed in skeletal muscle, although it was originally minus of smMLCK, which are not present in either of the cloned from cardiac muscle.16 skMLCK is reported to weakly striated muscle MLCKs, bind to thin actin filaments10 through bind to ,10 likely because it lacks the actin bind- an extension of the catalytic core toward the myosin thick fila- ing domain that is found in smMLCK (Figure 2). skMLCK ments for the phosphorylation of the smooth muscle MLC2 at phosphorylates skeletal muscle MLC2 (skMLC2) at Ser-15.17 Ser-19 (Figure 2).12 smMLCK-induced Ser-19 phosphoryla- In contrast to smMLCK, which phosphorylates only smooth tion activates myosin ATPase and initiates muscle contrac- muscle MLC2 efficiently, skMLCK can phosphorylate other tion.3 Interestingly, different kinases are also known to phos- MLC2 s found in cardiac and smooth muscles that exhibit phorylate smMLC2. Rho-associated coiled-coil forming kinase similar catalytic properties.17 Previous physiological studies (ROCK) phosphorylates smMLC2 at Ser-19 to regulate the have demonstrated that the extent of MLC2 phosphorylation assembly of stress fibers.13 Protein kinase C (PKC) phosphory- in skeletal muscle increases from 0–10% to 40–60% depend- lates Ser-1/Ser-2/Thr-9, which inhibits myosin ATPase activ- ing on the frequency of muscle stimulation.17 ity.14 In contrast, several protein kinases, including PKA, PKC, CaMKII, and PAK, are reported to phosphorylate serine resi- cMLCK dues in the calmodulin-binding sequence in the regulatory cMLCK is encoded by the MYLK3 gene and expressed ex- domain in vitro, which results in a 10-fold increase in Kcam.15 clusively in the heart, both the atria and ventricles.18 cMLCK On the other hand, the phosphorylation of smMLCK at Thr-40 is structurally related to both skMLCK and smMLCK and and Thr-43 by extracellular signal-regulated kinase (ERK) contains a conserved kinase domain at its C-terminus that increases Vmax without changing KCaM.15 ATP, which is the exhibits 58% identity with skMLCK and 44% identity with other substrate of MLCK, can bind to the MLCK catalytic core smMLCK.7,18 However, the N-terminus of cMLCK lacks ho- regardless of the positioning of the autoinhibitory sequence.12 mologies to known proteins, including other MLCKs, which Km for ATP is approximately 50–150 μmol/L.12 The concentra- indicates that cMLCK may play some specific functional roles tions of smMLCK and its substrate, smMLC2, are approxi- (Figure 2). Immunostaining of endogenous cMLCK in car- mately 4 μmol/L and 30–40 μmol/L, respectively.8 diac myocytes has shown a diffuse positive staining pattern in Advance Publication by-J-STAGE TSUKAMOTO O et al.

Table. Biochemical Characteristics of Protein Kinases in the Heart

Vmax Kinase Gene Substrate Km (μmol/L) Vmax/Km (μmol · min–1 · mg–1) Cardiac MLCK MYLK3 MLC2v 4.3±1.5 0.26±0.06 0.06 Skeletal MLCK MYLK2 Skeletal MLC2 4.3±0.5 40±1.7 9.3 Smooth muscle MLCK MYLK1 Smooth muscle MLC2 8.3±1.4 28±5.8 3.5 ZIPK DAPK3 MLC2v 15.2±2.0 0.89±0.05 0.06 Smooth muscle MLC2 1.8±0.3 0.42±0.03 0.23 MLCK, myosin light chain kinase; MLC2v, ventricular myosin regulatory chain-2; ZIPK, zipper-interacting protein kinase.

Figure 3. Physiological and biochemical effects of cardiac myosin light chain 2 (MLC2) phosphorylation by cardiac myosin light chain kinase (cMLCK) on cardiac muscle contraction. cMLCK-induced MLC2 phosphorylation enhances the contraction of car- diac myocytes probably via a combination of increased Ca2+ sensitization, myosin Mg-ATPase activity, and sarcomere structural organization. CaM, calmodulin. For explanations, see text.

the cytoplasm with a striated staining pattern in the cell pe- MLCK and skMLCK. However, the low specific activity of riphery.18 Interestingly, the striated MLCK staining colocal- cMLCK results in a slow turnover of phosphate in MLC2 ized with actin but not with its substrate.18 Chan et al reported (t1/2=250 min), with an MLC2 basal phosphorylation of ap- the independence of cMLCK activity from Ca2+/calmodulin,18 proximately 0.2–0.4 mol of phosphate/mol of MLC2 under which was an unexpected result, because cMLCK also con- basal conditions,23–26 which indicates that the kinase activity tains both autoinhibitory and calmodulin-binding sequences, of cMLCK may be a primary limiting factor of MLC2 phos- similar to the other 2 types of MLCKs, and binds to calmodu- phorylation.19 lin with high affinity in a Ca2+-dependent manner.7,19 There 2+ are conflicting results concerning the Ca /calmodulin depen- Structural Changes of Myosin Head Induced by dency of cMLCK activity,7,19 and thus requires further ex- amination. MLC2 Phosphorylation in Striated Muscles The Table demonstrates the estimated kinetic constants of Muscle myosin is the molecular motor in the thick filament of each MLCK determined by Lineweaver-Burk plots.18,20–22 the sarcomere and is composed of 1 pair of myosin heavy cMLCK has a high affinity for its substrate, similar to sk- chains (MHC) and 2 pairs of myosin light chains (MLC): es- MLCK and smMLCK. However, the catalytic efficiency of sential MLC and regulatory MLC (ie, MLC2) (Figure 1). Both cMLCK, which is indicated by the Vmax/Km ratio, is lower than of the MLCs wrap around the neck region of the MHC. MLC2 that of skMLCK and smMLCK. Thus, the maximal specific is positioned at the S1–S2 junction of the MHC through its kinase activity of cMLCK is much lower than that of sm- binding to a 35-amino-acid IQ motif on the MHC.27 The MLC2 Advance Publication by-J-STAGE Cardiac-Specific MLCK contains a highly conserved serine that is phosphorylatable by phosphorylation in a dose-dependent manner and, conversely, MLCK and plays an important role in the activation and mod- knockdown of cMLCK by RNAi decreases MCL2v phos- ulation of myosin by fine-tuning the motion of the neck region phorylation in vitro.18 Scruggs et al identified 3 distinct charge of the MHC.10 The MLC2 also contains a Ca2+/Mg2+ binding variants of endogenous MLC2v in vivo in the mouse: unphos- site at its N-terminus from Asp-37 to Asp-48, located in the phorylated, singly phosphorylated, and doubly phosphorylated first helix-loop-helix motif, and the binding of divalent cation at Ser-14/Ser-15.42 In contrast, Ser-14 in murine MLC2v is alters the structural and contractile properties.28,29 The neck replaced by Asn in human MLC2. However, human MLC2 region of the myosin head has been proposed to act as a lever also has 3 distinct charge variants in vivo: unphosphorylated, arm. The phosphorylation of MLC2 at Ser-15 results in the singly phosphorylated at Ser-15, and deamidated Asn-14/ addition of a negative charge to the N-terminal region of phosphorylated Ser-15.42 Interestingly, the deamination of Asn MLC2, which induces the myosin head to swing out from a to Asp can create a negative charge similar to that obtained position close to the thick filament’s backbone toward the actin through phosphorylation.26 In addition, there is a spatial gradi- filament, and this structural change increases the rate through ent of MLC2v phosphorylation through the ventricular wall: which the myosin-actin interaction occurs and promotes force relatively low in the inner layer and high in the outer layer.16,43 generation at a given level of Ca2+ (Figures 1,3).30,31 Interest- This gradient, which may be caused by reduced activity of the ingly, several mutations around the phosphorylatable Ser-15 phosphatase in the outer layer,44,45 may be important for nor- and the Ca2+ binding site in MLC2 have been found in patients malizing wall stress and contributes to efficient contraction of with familial hypertrophic cardiomyopathy.32–34 the whole heart.16 The level of MLC2v phosphorylation is maintained rela- PKs and Protein Phosphatase Regulation of tively constant by the appropriate balance between phosphor- ylation by cMLCK and dephosphorylation by phosphatase in MLC2 Phosphorylation in the Heart the physiologically constant beating heart. The dephosphory- There are 2 types of cardiac MLC2: a ventricular myosin light lation of MLC2v is mediated mainly by catalytic subunit of chain-2 (MLC2v) and an atrium-specific form (MLC2a).35 All type 1 phosphatase (PP1c-δ) in concert with myosin phospha- 3 MLCKs are expressed in the heart. However, the amount of tase target subunit 2 (MYPT2).46 In contrast, MLC2 phos- skMLCK in the heart is too low to maintain cardiac MLC2 phorylation by MLCK is transient in both smooth and skeletal phosphorylation,36 and ablation of skMLCK has no effect on muscle cells, although they have different MLC2 phosphory- MLC2 phosphorylation in cardiac muscle,36 which indicates lation dynamics.19 MLC2 phosphorylation is rapidly dephos- that skMLCK does not play a major role in MLC2v phos- phorylated by robust myosin light chain phosphatase activity phorylation. The expression of smMLCK in the heart is also in smooth muscle cells, whereas MLC2 phosphorylation is 10- to 20-fold lower than that of cMLCK,18 and cardiac MLC2 prolonged and slowly dephosphorylated by the low myosin is not a good substrate for smMLCK.36 smMLCK may phos- light chain phosphatase activity in skeletal muscle cells. phorylate non-muscle cytoplasmic myosin II-B and plays an 37 important role in the heart. Thus, there are currently 2 can- Physiological Regulators of cMLCK didate kinases for MLC2 phosphorylation in cardiac myo- cytes: cMLCK7,18 and zipper-interacting PK (ZIPK).38 ZIPK Expression and Activity is a Ca2+-independent serine/threonine kinase that has been The expression of cMLCK is highly regulated by the cardiac implicated in apoptosis and is ubiquitously expressed, in- homeobox protein Nkx2-5 in neonatal cardiac myocytes.18 cluding in adult and neonatal cardiac myocytes.39 In permea- cMLCK mRNA increases during development to adult stages bilized smooth muscle cells, constitutively active ZIPK initi- and persists in the aged heart, whereas its protein level de- ates contraction through the phosphorylation of smooth muscle creases in the aged heart, which suggests an alternation in MLC2.40 Cardiac MLC2 was also identified as a biochemically post-transcriptional regulation.18 Exercise has been reported to favorable substrate for ZIPK through an unbiased substrate increase cMLCK expression and MLC2v phosphorylation.25 search with purified ZIPK on heart homogenates and was phos- Catalytic activity may be regulated by phosphorylation through phorylated at Ser-15 in vivo and in vitro by ZIPK.38 In fact, upstream kinases.47 cMLCK has several potential phosphory- Vmax for cardiac MLC2 is 2-fold greater than that obtained for lation sites for other kinases, such as PKA and PKC,18 al- smooth muscle MLC2 (887±47 and 415±49 nmol · min–1 · mg–1, though more studies are required to clarify the details of this respectively), whereas the Km of cardiac MLC2 is higher than mechanism. that of smooth muscle MLC2 (15.0±2.0 and 1.8±0.3 μmol/L, Hypertrophic agonists, such as α1- or β1-adrenergic stimu- respectively).38 However, ZIPK is considered not to be in- lation18,48,49 and angiotensin II, induce MLC2v phosphorylation volved in the basal phosphorylation of cardiac MLC2 in vivo,41 through MLCK activation in both cultured cardiac myocytes although knockdown of ZIPK in cardiac myocytes by siRNA and the adult heart in vivo.24 Other neurohumoral stimulators, decreased the extent of cardiac regulatory MLC phosphoryla- such as endothelin50 and prostanoid F receptors,51 have also tion by 34%.38 The physiological role of ZIPK-induced cardiac been reported to increase MLC2 phosphorylation and induce a MLC2 phosphorylation remains unknown. prominent increase in the contractile force in the heart. Neu- It is now well established that cMLCK is the predominant regulin, which activates ErbB receptor tyrosine kinase, has cardiac MLC2 kinase responsible for basal phosphorylation in been reported to enhance the expression of cMLCK with a vivo, because the ablation of cMLCK almost completely abol- concomitant increase in MLC2v phosphorylation and improved ishes the phosphorylation of MLC2v.7,18,25,41 Ding et al demon- cardiac performance after myocardial infarction in rats.52 strated that a partial reduction in the amount of cMLCK pro- +/neo tein in cMLCK hetero knockout mice (cMLCK ) resulted Role of MLC2 Phosphorylation in Sarcomere in a partial reduction in MLC2 phosphorylation in both ven- tricular and atrial muscles, and this reduction in MLC2 phos- Organization and Heart Development phorylation was proportional to the cMLCK expression level.41 The transfection of cardiac myocytes with skMLCK has re- In addition, the overexpression of cMLCK increases MLC2v sulted in the phosphorylation of MLC2v and led to a highly Advance Publication by-J-STAGE TSUKAMOTO O et al. organized sarcomeric pattern without induction of other hy- whereas the fibers from the TG-MLC2v(P-) mice did not ex- pertrophic phenotypes, such as the induction of fetal genes and hibit these increases.53 This suppressed performance of muscle an increase in cell size.24 Furthermore, the dominant-negative fibers with nonphosphorable MLC2v is consistent with the kinase-inactive form of MLCK completely prevents sarco- demonstrated effects of MLC2 phosphorylation in skeletal mere organization in response to angiotensin II,24 which sug- muscle.36 Scruggs et al48 examined the role of regulatory myo- gests that MLCK activation is necessary and sufficient to in- sin light chain 2 phosphorylation in the ejection of the hearts duce sarcomere organization (Figure 3). Consistent with these of TG-MLC2v(P-) mice by measuring the systolic mechan- findings, the adenovirus-mediated overexpression of cMLCK ics under basal conditions and in response to adrenergic stim- in cardiac myocytes promotes sarcomere organization charac- ulation. The TG-MLC2v(P-) mice demonstrated depressed terized by straight, thick, striated actin bundles,18 whereas re- contractility, decreased maximal left ventricular power de- establishment of the phenylephrine-induced sarcomere struc- velopment, and a decrease in the time-to-peak elastance dur- ture is inhibited by pretreatment with RNAi against cMLCK.7 ing ejection under basal conditions.48 Interestingly, the TG- Interestingly, the reduced cMLCK expression induced by the MLC2v(P-) mice exhibited a blunting of the positive inotropic antisense morpholino causes severely impaired heart develop- response to β1-adrenergic stimulation.48 Because cMLCK has ment in zebrafish and histological analysis showed that the multiple PKA consensus sequences in its unique N-terminus structure of the sarcomere was poorly developed compared region,26 and β1-adrenergic stimulation increased MLC2v Ser- with control zebrafish.7 However, studies with transgenic and 15 phosphorylation in hearts of non-transgenic control mice48, knockout mice have shown that MLC2v phosphorylation is there might be a possible relationship between β1-adrenergic not critical for cardiogenesis in the mammalian system,18,41,53 signaling and MLC2v phosphorylation. The TG-MLC2v(P-) although it is necessary for the optimal contractile perfor- mice developed cardiac hypertrophy at 3–4 months of age, mance of the heart. most likely because of a compensatory hypertrophic growth response to diminished contractile performance.53 cMLCK- neo/neo Physiological Role of MLC2 knockout mice (cMLCK ) also demonstrated the critical role of cMLCK in normal physiological cardiac function, with Phosphorylation in the Heart decreased cardiac performance and the induction of cardiac The degree of MLC2 phosphorylation is known to play a criti- hypertrophy at 4–5 months of age.41 In contrast, transgenic cal role in the determination of the Ca2+-sensitive cross-bridge mice overexpressing MYPT2 specifically in cardiac myocytes transition in skeletal muscle.54 In permeable cardiac muscle demonstrated enhanced expression of MYPT2 with a con- fibers, MLC2 phosphorylation induced by MLCK increases comitant increase in the level of endogenous PP1c-δ, which the Ca2+ sensitivity, which manifests as a leftward shift in resulted in a reduction of the level of in vivo MLC2v phos- the force-Ca2+ relationship, and MLC2 dephosphorylation by phorylation in the heart associated with a decrease in the phosphatase decreases the Ca2+ sensitivity (ie, dephosphoryla- response to Ca2+ and a decreased left ventricular tion resulted in Ca2+ desensitization).55,56 MLC2 phosphoryla- contractility.61 These findings are consistent with the evidence tion by MLCK is also associated with enhanced Ca2+-stimu- obtained from TG-MLC2v(P-) mice and cMLCK-knockout lated myosin Mg-ATPase activity in rat cardiac myofibrils mice as mentioned earlier. (Figure 3).57 In the rat heart, MLC2v phosphorylation increas- Surprisingly, Warren et al recently reported an inverse rela- es in response to an increase in the beat frequency and/or left tionship between cMLCK expression and systolic pressure: ventricular pressure by exercise or inotropic agents, which may cMLCK expression is higher in the right ventricular myocar- help augment the peak left ventricular pressure.58,59 At the cel- dium than the left ventricular myocardium.25 Because the lular level, adenovirus-mediated overexpression of cMLCK cMLCK expression in the left ventricle was markedly down- potentiates the amplitude of the contraction of cardiac myo- regulated as a result of pressure overload, those authors spec- cytes and the kinetics of contraction and relaxation without ulated that increased mechanical stress reduces the net expres- changing the [Ca2+]i transients.18 sion of cMLCK.25 Several studies have also addressed the role of MLC2 phos- phorylation in the heart using genetically modified mice. Huang et al60 generated transgenic mice expressing skMLCK cMLCK in Heart Diseases specifically in cardiac myocytes (TG-skMLCK). These TG- cMLCK was first identified through the integrated cDNA ex- skMLCK mice demonstrated marked increases in the phos- pression analysis of failing human myocardium, which showed phorylation of both cardiac MLC2 and cytoplasmic non-mus- that the cMLCK mRNA expression levels correlated well with cle MLC2 in the heart without significant cardiac hypertrophy the pulmonary arterial pressure of patients with heart failure.7 or structural abnormalities up to 6 months of age, which in- Decreased phosphorylation of MLC2 was also reported in dicates that increased cardiac MLC2 phosphorylation per se some patients with heart failure.62,63 In animal models of myo- does not cause cardiac hypertrophy.60 Interestingly, the hyper- cardial infarction, the expression level of cMLCK in the heart trophic cardiac response to exercise and isoproterenol treat- has been found to be reduced.18,52 In addition, pressure over- ment was attenuated in TG-skMLCK mice,26 which supports load also led to a marked reduction in cMLCK and phosphory- the hypothesis that the phosphorylation of cardiac MLC2 may lated MLC2v in the heart 1 week after thoracic aortic constric- inhibit physiological and pathophysiological hypertrophic re- tion surgery.25 Interestingly, the reduction in the cMLCK sponses through enhanced contractile performance and effi- protein level in pressure-overloaded hearts was mediated by ciency. Sanbe et al53 created transgenic mice (TG-MLC2v(P-)) upregulation of the ubiquitin-proteasome degradation system.25 in which 3 potentially phosphorylatable serines (Ser-14/Ser- The specific overexpression of cMLCK in cardiac myocytes 15/Ser-19) in the MLC2v (ventricular regulatory myosin light attenuated the phenotype of the pressure overload-induced chain) were mutated to alanine. After MLCK treatment, the heart failure,25 and this finding suggests a protective role of isolated ventricular fibers from the non-transgenic control cMLCK against cardiac stress. mice showed increased Mg-ATPase activity and Ca2+ sensitiv- Induced pluripotent stem cell-derived cardiac myocytes ity, as indicated by a leftward shift in the force-Ca2+ curve, (iPSC-CMs) are expected to become a new cell therapy for Advance Publication by-J-STAGE Cardiac-Specific MLCK heart diseases and iPSC-CMs express cardiac-specific proteins 12. Hong F, Haldeman BD, Jackson D, Carter M, Baker JE, Cremo CR. similar to neonatal cardiac myocytes.64 However, so far there Biochemistry of smooth muscle myosin light chain kinase. Arch are no reports of investigations into cMLCK and MLC2 phos- Biochem Biophys 2011; 510: 135 – 146. 13. Totsukawa G, Yamakita Y, Yamashiro S, Hartshorne DJ, Sasaki Y, phorylation in iPSC-CMs. Matsumura F. Distinct roles of ROCK (Rho-kinase) and MLCK in spatial regulation of MLC phosphorylation for assembly of stress fibers and focal adhesions in 3T3 fibroblasts. J Cell Biol 2000; 150: New Approach to Finding Potential 797 – 806. Inhibitors or Activators of Kinases 14. Ikebe M, Hartshorne DJ, Elzinga M. Phosphorylation of the 20,000- dalton light chain of smooth muscle myosin by the calcium-activat- Recently, Suga et al developed a new technology to discover ed, phospholipid-dependent protein kinase: Phosphorylation sites “natural product-like” nonstandard peptides against various and effects of phosphorylation. J Biol Chem 1987; 262: 9569 – 9573. therapeutic targets, and this technology, which is known as the 15. Kamm KE, Stull JT. Dedicated myosin light chain kinases with di- verse cellular functions. J Biol Chem 2001; 276: 4527 – 4530. RaPID (Random non-strand Peptides Integrated Discovery) 16. Davis JS, Hassanzadeh S, Winitsky S, Lin H, Satorius C, Vemuri R, system, comprises a FIT (Flexible in vivo Translation) system et al. The overall pattern of cardiac contraction depends on a spatial coupled with an mRNA display.65 Using this system, Suga et gradient of myosin regulatory light chain phosphorylation. Cell 2001; al successfully isolated anti-E6AP macrocyclic N-methyl-pep- 107: 631 – 641. 17. Stull JT, Kamm KE, Vandenboom R. Myosin light chain kinase and tides, one of which had high affinity (Kd=0.60 nmol/L) against the role of myosin light chain phosphorylation in skeletal muscle. a target E6AP and strongly suppressed its polyubiquitination Arch Biochem Biophys 2011; 510: 120 – 128. ability against p53.66 Because the targets of the RaPID system 18. Chan JY, Takeda M, Briggs LE, Graham ML, Lu JT, Horikoshi N, are not limited, this technology can be applied to the discovery et al. Identification of cardiac-specific myosin light chain kinase. of kinase activators or inhibitors. Circ Res 2008; 102: 571 – 580. 19. Kamm KE, Stull JT. Signaling to myosin regulatory light chain in sarcomeres. J Biol Chem 2011; 286: 9941 – 9947. 20. Zhi G, Herring BP, Stull JT. Structural requirements for phosphory- Conclusion lation of myosin regulatory light chain from smooth muscle. J Biol We are now attempting to find a new potential activator or Chem 1994; 269: 24723 – 24727. 21. Herring BP, Gallagher PJ, Stull JT. Substrate specificity of myosin inhibitor of cMLCK, which will hopefully either provide new light chain kinases. J Biol Chem 1992; 267: 25945 – 25950. insights to cMLCK or be successfully applied in the clinical 22. Ikebe M, Reardon S, Schwonek JP, Sanders CR 2nd, Ikebe R. Struc- setting. tural requirement of the regulatory light chain of smooth muscle myo- sin as a substrate for myosin light chain kinase. J Biol Chem 1994; 269: 28165 – 28172. Acknowledgments 23. Herring BP, England PJ. The turnover of phosphate bound to myosin This research was supported by the Japan Society for the Promotion of light chain-2 in perfused rat heart. Biochem J 1986; 240: 205 – 214. Science (JSPS) through the “Funding Program for Next Generation World- 24. Aoki H, Sadoshima J, Izumo S. Myosin light chain kinase mediates Leading Researchers (NEXT Program),” initiated by the Council for Sci- sarcomere organization during cardiac hypertrophy in vitro. Nat Med ence and Technology Policy (CSTP); grants-in-aid from the Ministry of 2000; 6: 183 – 188. Health, Labor, and Welfare-Japan; grants-in-aid from the Ministry of Edu- 25. Warren SA, Briggs LE, Zeng H, Chuang J, Chang EI, Terada R, et cation, Culture, Sports, Science, and Technology-Japan. The research was al. Myosin light chain phosphorylation is critical for adaptation to also supported by grants from the Japan Heart Foundation, Japan Cardio- cardiac stress. Circulation 2012; 126: 2575 – 2588. vascular Research Foundation, Japan Cardiovascular Research Founda- 26. Scruggs SB, Solaro RJ. The significance of regulatory light chain tion, Japan Foundation of Applied Enzymology. phosphorylation in cardiac physiology. Arch Biochem Biophys 2011; Conflict of Interest: The authors have no conflicts of interest to declare. 510: 129 – 134. 27. Rayment I, Rypniewski WR, Schmidt-Base K, Smith R, Tomchick DR, Benning MM, et al. Three-dimensional structure of myosin References subfragment-1: A molecular motor. Science 1993; 261: 50 – 58. 1. Huxley H, Hanson J. Changes in the cross-striations of muscle dur- 28. Reinach FC, Nagai K, Kendrick-Jones J. Site-directed mutagenesis ing contraction and stretch and their structural interpretation.1954. of the regulatory light-chain Ca2+/Mg2+ binding site and its role in 2. Ebashi F, Ebashi S. Removal of calcium and relaxation in actomyo- hybrid . Nature 1986; 322: 80 – 83. sin systems. Nature 1962; 194: 378 – 379. 29. Szczesna-Cordary D, Guzman G, Ng SS, Zhao J. Familial hypertro- 3. Kamm KE, Stull JT. The function of myosin and myosin light chain phic cardiomyopathy-linked alterations in Ca2+ binding of human kinase phosphorylation in smooth muscle. Annu Rev Pharmacol cardiac myosin regulatory light chain affect cardiac muscle contrac- Toxicol 1985; 25: 593 – 620. tion. J Biol Chem 2004; 279: 3535 – 3542. 4. Koizumi K, Hoshiai M, Ishida H, Ohyama K, Sugiyama H, Naito A, 30. Levine RJ, Chantler PD, Kensler RW, Woodhead JL. Effects of phos- et al. Stanniocalcin 1 prevents cytosolic Ca2+ overload and cell hy- phorylation by myosin light chain kinase on the structure of Limulus percontracture in cardiomyocytes. Circ J 2007; 71: 796 – 801. thick filaments. J Cell Biol 1991; 113: 563 – 572. 5. Ebashi S, Kodama A. A new protein factor promoting aggregation 31. Levine RJ, Kensler RW, Yang Z, Stull JT, Sweeney HL. Myosin light of tropomyosin. J Biochem 1965; 58: 107 – 108. chain phosphorylation affects the structure of rabbit skeletal muscle 6. Kajioka S, Takahashi-Yanaga F, Shahab N, Onimaru M, Matsuda M, thick filaments. Biophys J 1996; 71: 898 – 907. Takahashi R, et al. Endogenous cardiac troponin T modulates Ca(2+)- 32. Szczesna D, Ghosh D, Li Q, Gomes AV, Guzman G, Arana C, et al. mediated smooth muscle contraction. Sci Rep 2012; 2: 979. Familial hypertrophic cardiomyopathy mutations in the regulatory 7. Seguchi O, Takashima S, Yamazaki S, Asakura M, Asano Y, Shintani light chains of myosin affect their structure, Ca2+ binding, and phos- Y, et al. A cardiac myosin light chain kinase regulates sarcomere phorylation. J Biol Chem 2001; 276: 7086 – 7092. assembly in the vertebrate heart. J Clin Invest 2007; 117: 2812 – 2824. 33. Otsuka H, Arimura T, Abe T, Kaiwai H, Aizawa Y, Kubo T, et al. 8. Takashima S. Phosphorylation of myosin regulatory light chain by Prevalence and distribution of sarcomeric gene mutations in Japanese myosin light chain kinase, and muscle contraction. Circ J 2009; 73: patients with familial hypertrophic cardiomyopathy. Circ J 2012; 76: 208 – 213. 453 – 461. 9. Chin D, Means AR. Calmodulin: A prototypical calcium sensor. 34. Niwano S. Multicenter study of the prevalence and distribution of Trends Cell Biol 2000; 10: 322 – 328. sarcometric gene mutations in familial hypertrophic cardiomyopa- 10. Sweeney HL, Bowman BF, Stull JT. Myosin light chain phosphory- thy: A milestone for genetic diagnosis in the Japanese population. lation in vertebrate striated muscle: Regulation and function. Am J Circ J 2012; 76: 303 – 304. Physiol 1993; 264: C1085 – C1095. 35. Collins JH. Myoinformatics report: Myosin regulatory light chain 11. Birukov KG, Schavocky JP, Shirinsky VP, Chibalina MV, Van Eldik paralogs in the human genome. J Muscle Res Cell Motil 2006; 27: LJ, Watterson DM. Organization of the genetic locus for chicken 69 – 74. myosin light chain kinase is complex: Multiple proteins are encoded 36. Zhi G, Ryder JW, Huang J, Ding P, Chen Y, Zhao Y, et al. Myosin and exhibit differential expression and localization. J Cell Biochem light chain kinase and myosin phosphorylation effect frequency- 1998; 70: 402 – 413. dependent potentiation of skeletal muscle contraction. Proc Natl Advance Publication by-J-STAGE TSUKAMOTO O et al.

Acad Sci USA 2005; 102: 17519 – 17524. 52. Gu X, Liu X, Xu D, Li X, Yan M, Qi Y, et al. Cardiac functional 37. Ma X, Takeda K, Singh A, Yu ZX, Zerfas P, Blount A, et al. Condi- improvement in rats with myocardial infarction by up-regulating tional ablation of nonmuscle myosin II-B delineates heart defects in cardiac myosin light chain kinase with neuregulin. Cardiovasc Res adult mice. Circ Res 2009; 105: 1102 – 1109. 2010; 88: 334 – 343. 38. Chang AN, Chen G, Gerard RD, Kamm KE, Stull JT. Cardiac myo- 53. Sanbe A, Fewell JG, Gulick J, Osinska H, Lorenz J, Hall DG, et al. sin is a substrate for zipper-interacting protein kinase (ZIPK). J Biol Abnormal cardiac structure and function in mice expressing non- Chem 2010; 285: 5122 – 5126. phosphorylatable cardiac regulatory myosin light chain 2. J Biol 39. Haystead TA. ZIP kinase, a key regulator of myosin protein phos- Chem 1999; 274: 21085 – 21094. phatase 1. Cell Signal 2005; 17: 1313 – 1322. 54. Patel JR, Diffee GM, Moss RL. Myosin regulatory light chain mod- 40. Ihara E, MacDonald JA. The regulation of smooth muscle contractil- ulates the Ca2+ dependence of the kinetics of tension development in ity by zipper-interacting protein kinase. Can J Physiol Pharmacol skeletal muscle fibers. Biophys J 1996; 70: 2333 – 2340. 2007; 85: 79 – 87. 55. Morano I, Hofmann F, Zimmer M, Ruegg JC. The influence of P- 41. Ding P, Huang J, Battiprolu PK, Hill JA, Kamm KE, Stull JT. Car- light chain phosphorylation by myosin light chain kinase on the diac myosin light chain kinase is necessary for myosin regulatory calcium sensitivity of chemically skinned heart fibres. FEBS Lett light chain phosphorylation and cardiac performance in vivo. J Biol 1985; 189: 221 – 224. Chem 2010; 285: 40819 – 40829. 56. Sweeney HL, Stull JT. Phosphorylation of myosin in permeabilized 42. Scruggs SB, Reisdorph R, Armstrong ML, Warren CM, Reisdorph mammalian cardiac and skeletal muscle cells. Am J Physiol 1986; N, Solaro RJ, et al. A novel, in-solution separation of endogenous 250: C657 – C660. cardiac sarcomeric proteins and identification of distinct charged 57. Noland TA Jr, Kuo JF. Phosphorylation of cardiac myosin light chain variants of regulatory light chain. Mol Cell Proteomics 2010; 9: 2 by protein kinase C and myosin light chain kinase increases 1804 – 1818. Ca(2+)-stimulated actomyosin MgATPase activity. Biochem Biophys 43. Hidalgo C, Wu Y, Peng J, Siems WF, Campbell KB, Granzier H. Res Commun 1993; 193: 254 – 260. Effect of diastolic pressure on MLC2v phosphorylation in the rat left 58. Fitzsimons DP, Bodell PW, Baldwin KM. Phosphorylation of rodent ventricle. Arch Biochem Biophys 2006; 456: 216 – 223. cardiac myosin light chain 2: Effects of exercise. J Appl Physiol 44. Rajashree R, Blunt BC, Hofmann PA. Modulation of myosin phos- 1989; 67: 2447 – 2453. phatase targeting subunit and protein phosphatase 1 in the heart. Am 59. Fitzsimons DP, Bodell PW, Baldwin KM. Myocardial functional J Physiol Heart Circ Physiol 2005; 289: H1736 – H1743. correlates of cardiac myosin light chain 2 phosphorylation. J Appl 45. Cohen PT. Protein phosphatase 1: Targeted in many directions. J Physiol 1990; 68: 2426 – 2433. Cell Sci 2002; 115: 241 – 256. 60. Huang J, Shelton JM, Richardson JA, Kamm KE, Stull JT. Myosin 46. Fujioka M, Takahashi N, Odai H, Araki S, Ichikawa K, Feng J, et al. regulatory light chain phosphorylation attenuates cardiac hypertro- A new isoform of human myosin phosphatase targeting/regulatory phy. J Biol Chem 2008; 283: 19748 – 19756. subunit (MYPT2): cDNA cloning, tissue expression, and chromo- 61. Mizutani H, Okamoto R, Moriki N, Konishi K, Taniguchi M, Fujita somal mapping. Genomics 1998; 49: 59 – 68. S, et al. Overexpression of myosin phosphatase reduces Ca(2+) sen- 47. Soderling TR, Stull JT. Structure and regulation of calcium/calmod- sitivity of contraction and impairs cardiac function. Circ J 2010; 74: ulin-dependent protein kinases. Chem Rev 2001; 101: 2341 – 2352. 120 – 128. 48. Scruggs SB, Hinken AC, Thawornkaiwong A, Robbins J, Walker 62. Morano I. Effects of different expression and posttranslational mod- LA, de Tombe PP, et al. Ablation of ventricular myosin regulatory ifications of myosin light chains on contractility of skinned human light chain phosphorylation in mice causes cardiac dysfunction in situ cardiac fibers. Basic Res Cardiol 1992; 87(Suppl 1): 129 – 141. and affects neighboring myofilament protein phosphorylation. J Biol 63. van der Velden J, Papp Z, Boontje NM, Zaremba R, de Jong JW, Chem 2009; 284: 5097 – 5106. Janssen PM, et al. The effect of myosin light chain 2 dephosphoryla- 49. Grimm M, Haas P, Willipinski-Stapelfeldt B, Zimmerman WH, Rau tion on Ca2+-sensitivity of force is enhanced in failing human hearts. T, Pantel K, et al. Key role of myosin light chain (MLC) kinase- Cardiovasc Res 2003; 57: 505 – 514. mediated MLC2a phosphorylation in the alpha 1-adrenergic positive 64. Yu T, Miyagawa S, Miki K, Saito A, Fukushima S, Higuchi T, et al. inotropic effect in human atrium. Cardiovasc Res 2005; 65: 211 – In vivo differentiation of induced pluripotent stem cell-derived car- 220. diomyocytes. Circ J 2013; 77: 1297 – 1306. 50. Rossmanith GH, Hoh JF, Turnbull L, Ludowyke RI. Mechanism of 65. Hipolito CJ, Suga H. Ribosomal production and in vitro selection of action of endothelin in rat cardiac muscle: Cross-bridge kinetics and natural product-like peptidomimetics: The FIT and RaPID systems. myosin light chain phosphorylation. J Physiol 1997; 505: 217 – 227. Curr Opin Chem Biol 2012; 16: 196 – 203. 51. Riise J, Nguyen CH, Qvigstad E, Sandnes DL, Osnes JB, Skomedal 66. Yamagishi Y, Shoji I, Miyagawa S, Kawakami T, Katoh T, Goto Y, T, et al. Prostanoid F receptors elicit an inotropic effect in rat left et al. Natural product-like macrocyclic N-methyl-peptide inhibitors ventricle by enhancing myosin light chain phosphorylation. Cardio- against a ubiquitin ligase uncovered from a ribosome-expressed de vasc Res 2008; 80: 407 – 415. novo library. Chem Biol 2011; 18: 1562 – 1570.